lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aroop Ganguly <>
Subject Re: Solr Odbc for Parallel Sql integration with Tableau
Date Tue, 19 Jun 2018 06:14:29 GMT
Hi Joel

Yes I was able to make the ODBC bridge work very easily (using steps mentioned here
<> ), 

But the actual Tableau integration has not been fruitful yet due to 2 reasons:

1. Tableau inherently writes inner queries : select a as A, b as B from (select * from c)
     — this fails immediately, as parallel sql in my experience does not like inner queries.

2. The default Tableau view which is awesome to drag and drop the entire table does not work
for parallel sql as we need to specify a “limit” otherwise it keeps giving the error about
“score”. So I defaulted to the custom query option on Tableau but it failed for the inherent
inner-queryness of Tableau as mentioned in 1. :) 
 — I will keep at it tomorrow and maybe I will be able to figure a way out.

> On Jun 18, 2018, at 7:55 PM, Joel Bernstein <> wrote:
> That's interesting that you were able to setup OpenLink. At Alfresco we've
> done quite a bit of work on the Solr's JDBC driver to integrate it with the
> Alfresco repository, which uses Solr. But we haven't yet tackled the ODBC
> setup. That will come very soon. To really take advantage of Tableau's
> capabilities we will need to add joins to Solr's parallel SQL. Solr already
> uses Apache Calcite, which has a join optimizer, so mainly this would
> involve hooking up the various Streaming Expression joins.
> Joel Bernstein
> On Mon, Jun 18, 2018 at 6:37 PM, Aroop Ganguly <>
> wrote:
>> Ok I was able to setup the odic bridge (using OpenLink) and I see the
>> collections popping up in Tableau too.
>> But I am unable to actually get data flowing into Tableau reports because,
>> Tableau keeps creating inner queries and Solr seems to hate inner queries.
>> Is there a way to do inner queries in Solr Parallel Sql ?
>>> On Jun 18, 2018, at 12:30 PM, Aroop Ganguly <>
>> wrote:
>>> Hi Everyone
>>> I am not sure if something has been done on this yet, though I did see a
>> JIRA with links to the parallel sql documentation, but I do not think that
>> answers the question.
>>> I love the jdbc driver and it works well for many UIs but there are
>> other systems that need an ODBC driver.
>>> Can anyone share any guidance as to how this can be done or has been
>> done by others.
>>> Thanks
>>> Aroop

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message